Re: 2.6.23-mm1

[Jeff Garzik <jeff@garzik.org>]

Torsten Kaiser wrote:
> On 10/13/07, Jeff Garzik <jeff@garzik.org> wrote:
>> Torsten Kaiser wrote:
>>> On 10/12/07, Andrew Morton <akpm@linux-foundation.org> wrote:
>>>> On Fri, 12 Oct 2007 10:31:42 +0200 "Torsten Kaiser" <just.for.lkml@googlemail.com> wrote:
>>>>> Oct 12 10:23:03 treogen smartd[6091]: Device: /dev/sdc, not found in
>>>>> smartd database.
>>>> hm.
>>>>
>>>>> Oct 12 10:23:03 treogen [  105.990000] WARNING: at
>>>>> drivers/ata/libata-core.c:5752 ata_qc_issue()
>>>> Let's cc linux-ide.
>>>>
>>>>> Oct 12 10:23:03 treogen [  105.990000]
>>>>> Oct 12 10:23:03 treogen [  105.990000] Call Trace:
>>>>> Oct 12 10:23:03 treogen [  105.990000]  [<ffffffff804442ef>]
>>>>> ata_qc_issue+0x47f/0x540
>>>>> Oct 12 10:23:03 treogen [  105.990000]  [<ffffffff80432e60>] scsi_done+0x0/0x20
>>>>> Oct 12 10:23:03 treogen [  105.990000]  [<ffffffff80449c80>]
>>>>> ata_scsi_flush_xlat+0x0/0x30
>>> Oct 13 07:46:48 treogen [   99.850000]
>>> Oct 13 07:46:48 treogen [   99.850000] ata3: EH in SWNCQ
>>> mode,QC:qc_active 0x3 sactive 0x1
>>> Oct 13 07:46:48 treogen [   99.850000] ata3: SWNCQ:qc_active 0x1
>>> defer_bits 0x0 last_issue_tag 0x0
>> The WARNING indicates that there is a SWNCQ bug in sata_nv.  Given that
>> the problem appears when SYNCHRONIZE CACHE is being issued, I would
> 
> I can't follow you on SYNCHRONIZE CACHE.
> The only command written to the syslog in the errors where
> 0x60==ATA_CMD_FPDMA_READ and 0xB0 (which is not in
> include/linux/ata.h, but ATA-6 says that this is SMART related. That
> makes sense, as smartd is failing).

In the traceback you have "ata_scsi_flush_xlat", which is the function 
that translates a SCSI sync-cache command into an ATA flush-cache command.

The "WARNING: at drivers/ata/libata-core.c:5752 ata_qc_issue()" also 
guides us to the code comment

         /* Make sure only one non-NCQ command is outstanding.  The
          * check is skipped for old EH because it reuses active qc to
          * request ATAPI sense.
          */

which is a check related to NCQ->off and off->NCQ edge cases.

So those are the two bits of information I found interesting.


>> guess that sata_nv is not properly handling non-queued commands.
> 
> But that still seems correct, as I would not expect that SMART
> commands get queued. (Thats just a guess, as I did not try to find the
> code that does this distinction)
> 
>> This is a patch from libata-dev.git#nv-swncq (via #ALL).
> 
> Comparing sata_nv.c from 2.6.23-rc8-mm1 and 2.6.23-mm1 I see two
> changes, that look suspicious:
> 
> http://git.kernel.org/?p=linux/kernel/git/jgarzik/libata-dev.git;a=commitdiff;h=31cc23b34913bc173680bdc87af79e551bf8cc0d
> 
> The comment says: "ahci and sata_sil24 are converted to use ata_std_qc_defer()."
> But the patch also adds ".qc_defer = ata_std_qc_defer," to sata_nv.c
> 
> The second change is the removal of the 'lock' spinlock from sata_nv.c
> that was used in nv_swncq_qc_issue and nv_swncq_host_interrupt.
> 
> Should I try to revert one or both of these changes?

If you are git-capable, IMO the next steps in problem elimination should be

* download latest linux-2.6.git (currently 
752097cec53eea111d087c545179b421e2bde98a)
* build and test linux-2.6.git, to establish a new baseline
* download latest libata-dev.git#nv-swncq (currently 
3cb664c2d319a4fde5028c3c5dab6221fe70bd2d)
* build and test, with sata_nv module option swncq=0
* build and test, with sata_nv module option swncq=1

That will get -mm out of the picture, use the same baseline kernel for 
all three tests (nv-swncq is based off of 
752097cec53eea111d087c545179b421e2bde98a) and narrow things down to the 
precise changes that went upstream (or are on the 'nv-swncq' branch, 
waiting to go upstream).

My gut feeling is that there is a lingering bug in sata_nv SWNCQ somewhere.

	Jeff